---
dataset_info:
  features:
  - name: image
    dtype: image
  - name: captions
    sequence: string
  - name: class
    dtype: string
  - name: family
    dtype: string
  splits:
  - name: train
    num_bytes: 52287734.0
    num_examples: 640
  - name: validation
    num_bytes: 13119705.0
    num_examples: 160
  download_size: 64509780
  dataset_size: 65407439.0
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: validation
    path: data/validation-*
---